WO2013054019A1 - Procédé de contrôle d'un robot possédant au moins deux jambes - Google Patents
Procédé de contrôle d'un robot possédant au moins deux jambes Download PDFInfo
- Publication number
- WO2013054019A1 WO2013054019A1 PCT/FR2012/052216 FR2012052216W WO2013054019A1 WO 2013054019 A1 WO2013054019 A1 WO 2013054019A1 FR 2012052216 W FR2012052216 W FR 2012052216W WO 2013054019 A1 WO2013054019 A1 WO 2013054019A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- control method
- optimization
- leg
- displacement
- robot
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/18—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form
- G05B19/19—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form characterised by positioning or contouring control systems, e.g. to control position from one programmed point to another or to control movement along a programmed continuous path
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B62—LAND VEHICLES FOR TRAVELLING OTHERWISE THAN ON RAILS
- B62D—MOTOR VEHICLES; TRAILERS
- B62D57/00—Vehicles characterised by having other propulsion or other ground- engaging means than wheels or endless track, alone or in addition to wheels or endless track
- B62D57/02—Vehicles characterised by having other propulsion or other ground- engaging means than wheels or endless track, alone or in addition to wheels or endless track with ground-engaging propulsion means, e.g. walking members
- B62D57/032—Vehicles characterised by having other propulsion or other ground- engaging means than wheels or endless track, alone or in addition to wheels or endless track with ground-engaging propulsion means, e.g. walking members with alternately or sequentially lifted supporting base and legs; with alternately or sequentially lifted feet or skid
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10—TECHNICAL SUBJECTS COVERED BY FORMER USPC
- Y10S—TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10S901/00—Robots
- Y10S901/01—Mobile robot
Definitions
- the invention is in the field of robotics, and more specifically in the field of robot control methods having a body supported by at least two legs on which they rely and progress.
- the period of steps has been fixed in one way or another. This is the case, for example, in the disclosure of the document "Online walking motion generation with automatic foot step placement", Herdt et al., Advanced robotics, 24, 719, 2010, which presents a use of the predictive command to generate a step for a biped robot based on a predefined reference speed, a free adaptation of the foot location and constraints on the foot pressure center of the robot on the ground.
- the robot model used is simple, but still to control the robot on the basis of a fixed period of step, some authors have developed more complicated models, including for example hip movements to account for a complex dynamic approaching that of a human being.
- the document US 2008/0133053 addresses the problem of stability of a robot undergoing thrust.
- the robot is modeled by a inverted pendulum including a free wheel to reproduce the effects of modification of angular moments of the body.
- the authors disclose a control method operating by the instantaneous determination of a position on the ground for a contact of a leg with the ground, and an angular momentum of the freewheel.
- Such a control method in fact it relates to a set of at least two steps, has the disadvantage that the calculation can be particularly long to conduct, which requires a greater computing power and also reduces the reactivity of the computer. system.
- the invention aims to solve these problems in particular, while decreasing the amount of calculations required.
- the invention consists of a method for controlling the movement on a floor of a device having at least two legs, comprising
- a step of optimization, possibly under constraints, of a displacement command for the device taking into account said current values and a set of displacement variables of the device, said optimization being carried out by modeling a displacement of the device between a current state and a second state, the device resting, during said movement on at least one first leg, and then from a moment of change of center of pressure on the ground, at least on a second leg,
- the modeling proposed in the invention is based on a characteristic state of the gait cycle corresponding to a maximum of potential energy.
- This state is the intersection of the path of the system with a Poincaré section, ie a subspace of the state space, characterized by a constraint on the state variables.
- the control method makes it possible to define control parameters for a biped robot in a precise and rapid manner, as such Transition model makes it possible to explain equations of motion by an analytical treatment, notably by using both the time and the length of steps as control variables.
- the proposed control method uses at least half as many control variables as that of the prior art as presented in the article by Mordatch et al.
- the Poincaré sections are well adapted to the study of the stability of the cyclic systems, and consequently the displacement modeling used by the invention is better suited to the analysis of the control than the modeling considered in the prior art. , for example in the article by Mordatch et al, which models the displacement between foot change states.
- the potential gravity energy of the device is adapted in the context of a model of the device.
- the movement control comprises a position on the ground for a contact of a leg with the ground or a time for a contact of a leg with the ground, or both.
- said optimization is performed using a linearized displacement model under the assumption that the displacement follows a polynomial trajectory, for example rectilinear or circular.
- the calculations are then considerably simplified.
- the optimization is performed using a linear quadratic controller, and alternatively, it is performed by model inversion.
- the optimization can be carried out as a function of a setpoint determined as a function of a continued trajectory, and / or as a function of a setpoint determined as a function of a continued speed .
- the optimization can be carried out under a constraint of respect of a cone of friction during a contact of a leg with the ground.
- the optimization can be performed according to a setpoint comprising at least one position of the device or a device speed and / or according to an average step length instruction, or a period (duration) of average step.
- the potential energy is a function of a distance between the center of mass and a point of contact with the ground of the leg in contact with the ground.
- the invention also consists of a humanoid or nonhumanoid robot incorporating a control device implementing a control method as mentioned above.
- Figures 1 and 2 show the robot and displacement models used.
- Figure 3 shows a general flowchart of a method according to the invention.
- Figure 4 shows a first embodiment of the invention.
- Figures 5 to 7 show results relating to the simulation of a first movement of a robot according to the first embodiment of the invention.
- Figure 8 shows a second embodiment of the invention.
- FIGS. 9 to 11 show results relating to the simulation of the same first displacement of a robot according to this time the second embodiment of the invention.
- FIGS. 12 to 14 show results relating to the simulation of a second displacement of a robot according to the second embodiment of the invention.
- Figure 15 shows a variant of the second embodiment of the invention.
- FIGS. 16 to 18 show results relating to the simulation of a third displacement of a robot according to the second embodiment of the invention.
- Figure 19 shows another variant of the second embodiment of the invention.
- Figures 20 to 22 show results relating to the simulation of a fourth movement of a robot according to the second embodiment of the invention.
- FIG 1 there is shown schematically the physical model of robot used in the invention.
- the robot 100 includes a center of mass 110, a leg 120 and a foot 130.
- the leg 120 connects the center of mass 110 and the foot 130.
- a second leg is also a component of the robot 100, but it is not shown in Figure 1. Indeed, for the mathematical model modeling the robot 100 which will be developed below, just consider a single leg 120.
- the robot 100 is modeled as an inverted linear pendulum. Assuming that one of the robot's feet remains permanently in contact with the ground, the system finds a simple mathematical expression in the form of the equation with the following notations
- This plane is determined by the speed and acceleration vectors of the center of mass at an initial instant.
- the potential energy of the robot 100 then takes the form being the characteristic pulsation of the system, depending on g and z c .
- the length of the legs is variable. This is represented as a telescopic leg 120 in Figure 1. Alternatively, the length of the legs may vary due to the presence of a knee.
- the index 0 indicates an initial value at time 0.
- the variable t represents the time.
- the vector U is a vector of dimension 2 defined by the positions of the center of mass 110 and the foot 130.
- the vector V of dimension 2 is representative of the speed of the center of mass in the horizontal plane, the latter being related to the pulsation.
- the duration between instants t 1 and t p is denoted Ti.
- the duration between the instants t p and t 2 is denoted T 2 .
- the transition model represented by FIG. 2 relates to the transition from a first state of the system 100 (instant ti) to a second state of the system 100, in which the potential energy is maximum (instant t 2 ), these two states being separated by a single foot exchange in contact with the ground (instant t p ).
- FIG. 1 represents the system 100 at the instant t 2 , that is to say in a state of maximum potential energy, whereas the system 100 moves along the trajectory T in the plane P.
- This method firstly comprises a step 300 for measuring the current values ⁇ of state variables. it can be done, according to different embodiments, using different sensors.
- the robot also decides in which plane P he wishes to change his center of mass 110, according to the current values ⁇ .
- a displacement setpoint is set for the system, and this system is during a step 310 of processing the setpoint by a tracking servo transformed into a set of state variables.
- the state variable setpoint ⁇ is processed by a controller based on the transition model presented in FIG. 2, during a step 320.
- the controller 320 produces control variables V which are used during a step 330 for controlling the robot 100 itself as a simple model or as instructions for a complex model of the system.
- FIG. 4 there is shown a first detailed embodiment of the invention. We have previously defined a trajectory to follow, in the form of a displacement instruction.
- the method firstly comprises a step 400 for measuring the current values ⁇ of state variables.
- a projection step 410 using the displacement instruction also allows the definition of a velocity vector V (state variable setpoint
- the projection of step 410 consists in defining the velocity vector (state variable setpoint ⁇ ) as a function of the difference between the current position of the robot 100 and a position considered as corresponding to this current position on the trajectory.
- the position of the robot 100 is taken as the average of the positions of the mass center of the robot 100 in two successive maximum potential energy states.
- the position along the trajectory corresponding to the position of the robot 100 is taken to be an orthogonal projection of the position of the robot 100 on the path.
- the projection is an orthogonal projection on the right.
- the projection on the circle is performed along a radius of the circle.
- the speed vector V and the current values ⁇ are then used during a control step 420 by a controller based on a linearized transition model, derived from the model presented above in equations 5.
- the servo-control step 420 performed by a controller based on the linearized transition model is performed by making linearization hypotheses according to which, according to one embodiment, the displacement of the robot 100 is quasi-rectilinear, and that its pace is symmetrical.
- step 420 the controller used performs local control of the linear quadratic regulator (LQR) type.
- LQR linear quadratic regulator
- the initialization of the controller 420 comprises a preliminary step 415 of calculating a nominal control Ti n0 m by an additional servo-control according to a desired pitch length means Step av - More specifically, the term is not calculated according to the current speed V 0 as follows With reference to FIGS. 5 to 7, the results of displacement of a robot 100 using the control method of FIG. 4 are represented.
- the pulsation is equal to 3 s "1.
- the duration of pitch Ti + T 2 is constrained between 0.2 and 0.8 s and the length of step is constrained between 10 cm and 1.2 m.
- Command variables and Ti constitute the control variables f.
- the mass of the robot is 70 kg.
- the following setpoint is defined: starting from a coordinate point (0; 0) a constant speed of 0.7 m. s "1 in the direction it is required to robot 100 for 3 s Then a straight line perpendicular to the direction of the previous speed must be reached and traveled at a speed of 0.7 ms -1 for 20 s.
- a lateral perturbation equivalent to 700 N for 0.2 s is applied when the robot 100 reaches the middle of the previous line.
- the robot is asked to traverse a circle with a radius of 2 m at a speed of 0.5 m. s "1 for 27 s.
- Figure 5 shows successive positions of the center of mass 110 and the foot 130. It is found that the lateral disturbance causes an overhang of about 3.5 m.
- Figure 6 are shown the horizontal speeds along the two axes.
- FIG. 7 is represented the duration of the step of the robot during its displacement, in the form of graphical representations of Ti and Ti + T 2 .
- a step 800 of measurement of current values of state variables ⁇ is first performed.
- Step 810 is similar to the projection step 410 of the embodiment relating to Figure 4.
- This projection step allows the calculation of a setpoint speed V, and a setpoint position vector U.
- These two state variables constituting a set of state variables
- the controller used is based on the transition model defined in Equations 4, in its exact form, and uses a model inversion algorithm.
- the optimization criterion for the model inversion is taken equal
- Ud and Vd are the desired state variables
- Qu and Qv are weighting matrices.
- the system of equations 5 is used as equalization constraints, and the duration of the step, the positioning of the foot and the characteristics of the soil can be used as constraints of inequalities.
- the extracted variables are the quantities ⁇ 1 , T 2 and the optimization is performed on a step of the robot 100.
- Command variables are then used during a step 830 for the control of a complex model of the robot 100 or the robot 100 itself.
- FIG. 9 represents the trajectory of the robot 100
- FIG. 10 represents its horizontal speeds along the two axes
- FIG. 11 represents the duration of the steps during the robot's journey.
- the set point is the same as that given in the scenario of FIGS. 5 to 7. It can be seen in FIG. 9 that the method using a model non-linear control allows a faster response to the disturbance, since the lateral overhang is less than 2 m, contrary to what is observed in Figure 5.
- FIG. 12 there is shown the trajectory of the robot.
- the inserts are enlargements of areas where the robot turns around or stops.
- Figure 13 shows the evolution of the speed along both axes over time.
- FIG. 14 represents the evolution of the step length as a function of time.
- FIG. 15 shows a control method according to the invention which differs from that represented in FIG. 8 by the introduction of constraints 1500 relating to control variables related to
- the robot moves in a staircase following a straight line at a speed of 0.4 m. s "1 while it is pushed forward by applying a disturbance equivalent to 175 N for 0.2 s, then laterally by applying a perturbation equivalent to 500 N for 0.2 s.
- Figures 16 to 18 show results obtained with the control method according to Figure 15.
- the step length is constrained 0.25 m in the direction y and the time ⁇ is constrained between 0.1 s and 0.8 s.
- the step duration may be shorter.
- Figure 16 shows the movement of the robot 100 in the plane. Inserts have been introduced to show that the constant pitch length constraint along the direction y is respected.
- FIG. 18 represents the duration of the robot's pitch as a function of time.
- FIG. 19 represents a control method of a robot according to the invention taking into account constraints 1900 relating to the state variables ⁇ . These constraints are here related to the characteristics of soil friction. They are introduced at controller 820.
- FIGS 20 to 22 show the results obtained with the method of Figure 19.
- the setpoint scenario given to the robot is as follows: it must move in the y direction of 0.5 m. s 1 .
- the environmental constraints are such that the coefficient of friction initially at the value of 1, is modified every 5 s. passing successively to the values 0.4, 0.3 and 0.2.
- Figure 20 shows the robot moving in the plane.
- Figure 21 shows the variation of speeds along both axes as a function of time.
- Figure 22 shows the variation of the step time as a function of time. The step accelerates when the coefficient of friction reaches the value 0.3.
- control of FIG. 8 is four times faster than the control of FIG. 11. This requires a calculation time of 25 ms per step of movement on a dual processor computer, the command having been programmed in Matlab language. (TM). Such results make it possible to consider the use of this command in an interactive context, or in a real-time context.
- a command based on the simultaneous use of the commands of FIGS. 4 and 8 is used.
- step 300 instead of measuring the values of the state variables in step 300, they are estimated based on different available information.
- control variables of a complex model are optimized by using the principles presented in the embodiments of the inverted linear pendulum model.
- a control based on the proposed modeling must consist of at least two free control variables in order to guarantee the stability of the simple system.
- the time Ti can be imposed if the position of the foot is free in both x and y.
- the time Ti and the position of the foot in the lateral direction must be free.
- control variable Ti is sufficient provided to consider at least two successive steps in the optimization calculation.
- the robot can be modeled as a compass-type system, that is with a center of mass connected to at least two feet by constant length legs.
- the robot may have more than two legs, for example four legs.
- it evolves with, permanently, a single leg raised and three legs lifted.
- it evolves with several legs laid and one or more legs lifted. In the latter case, between two leg changes, the center of pressure on the ground is immobile. Leg changes are therefore changes in the center of pressure on the ground.
- the concept of leg change covers, in addition to the laying of one leg and the simultaneous lifting of another leg, the laying of a leg without simultaneous lifting of another leg, and also the lifting of a leg without Simultaneous placement of another leg, provided that one leg remains in contact with the ground permanently.
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Human Computer Interaction (AREA)
- Manufacturing & Machinery (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Manipulator (AREA)
- Feedback Control In General (AREA)
Abstract
Description
Claims
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2014535136A JP6054404B2 (ja) | 2011-10-11 | 2012-10-01 | 少なくとも2本の脚を有するロボットの制御方法 |
EP12775799.5A EP2766251B1 (fr) | 2011-10-11 | 2012-10-01 | Procédé de contrôle d'un robot possédant au moins deux jambes |
US14/350,671 US9317027B2 (en) | 2011-10-11 | 2012-10-01 | Method of controlling a robot |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
FR1159152 | 2011-10-11 | ||
FR1159152A FR2981173B1 (fr) | 2011-10-11 | 2011-10-11 | Procede de controle d'un robot |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2013054019A1 true WO2013054019A1 (fr) | 2013-04-18 |
Family
ID=47071383
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/FR2012/052216 WO2013054019A1 (fr) | 2011-10-11 | 2012-10-01 | Procédé de contrôle d'un robot possédant au moins deux jambes |
Country Status (5)
Country | Link |
---|---|
US (1) | US9317027B2 (fr) |
EP (1) | EP2766251B1 (fr) |
JP (1) | JP6054404B2 (fr) |
FR (1) | FR2981173B1 (fr) |
WO (1) | WO2013054019A1 (fr) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9381645B1 (en) | 2014-12-08 | 2016-07-05 | Amazon Technologies, Inc. | System for automating manipulation of items |
CN112051797B (zh) * | 2020-09-07 | 2023-12-26 | 腾讯科技(深圳)有限公司 | 足式机器人运动控制方法、装置、设备及介质 |
CN114137954A (zh) * | 2021-10-14 | 2022-03-04 | 深圳鹏行智能研究有限公司 | 一种四足机器人质心轨迹规划方法、装置及存储介质 |
CN114434241B (zh) * | 2021-12-30 | 2023-06-02 | 西门子(中国)有限公司 | 控制石磨机横梁运动的方法、装置、电子设备及存储介质 |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080133053A1 (en) | 2006-11-29 | 2008-06-05 | Honda Motor Co., Ltd. | Determination of Foot Placement for Humanoid Push Recovery |
US20110029130A1 (en) * | 2005-03-30 | 2011-02-03 | Honda Motor Co., Ltd. | Systems and Methods for Controlling a Legged Robot Using a Two-Phase Disturbance Response Strategy |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1306792B1 (fr) * | 2001-10-29 | 2015-05-27 | Honda Giken Kogyo Kabushiki Kaisha | Système et procédé de simulation pour des dispositifs d'augmentation d'un humain |
JP4332393B2 (ja) * | 2003-09-16 | 2009-09-16 | 株式会社国際電気通信基礎技術研究所 | 2足歩行移動体および2足歩行制御方法 |
JP2007061954A (ja) * | 2005-08-31 | 2007-03-15 | Institute Of Physical & Chemical Research | 腰上下動による二足歩行ロボットとその歩行制御方法 |
JP2008080418A (ja) * | 2006-09-26 | 2008-04-10 | Yaskawa Electric Corp | 二足歩行ロボットの制御装置 |
EP2233387A1 (fr) * | 2009-03-24 | 2010-09-29 | Disney Enterprises, Inc. | Systèmes et procédés de suivi et d'équilibrage de robots pour imiter les données de capture en mouvement |
-
2011
- 2011-10-11 FR FR1159152A patent/FR2981173B1/fr active Active
-
2012
- 2012-10-01 JP JP2014535136A patent/JP6054404B2/ja not_active Expired - Fee Related
- 2012-10-01 WO PCT/FR2012/052216 patent/WO2013054019A1/fr active Application Filing
- 2012-10-01 US US14/350,671 patent/US9317027B2/en not_active Expired - Fee Related
- 2012-10-01 EP EP12775799.5A patent/EP2766251B1/fr not_active Not-in-force
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110029130A1 (en) * | 2005-03-30 | 2011-02-03 | Honda Motor Co., Ltd. | Systems and Methods for Controlling a Legged Robot Using a Two-Phase Disturbance Response Strategy |
US20080133053A1 (en) | 2006-11-29 | 2008-06-05 | Honda Motor Co., Ltd. | Determination of Foot Placement for Humanoid Push Recovery |
Non-Patent Citations (6)
Title |
---|
HERDT ET AL.: "Online walking motion generation with automatic foot step placement", ADVANCED ROBOTICS, vol. 24, 2010, pages 719 |
HYON; FUJIMOTO, IFAC SYMPOSIUM ON ROBOT CONTROL, 2009 |
IGOR MORDATCH ET AL: "Robust physics-based locomotion using low-dimensional planning", ACM TRANSACTIONS ON GRAPHICS, vol. 29, no. 4, 26 July 2010 (2010-07-26), pages 1, XP055035036, ISSN: 0730-0301, DOI: 10.1145/1778765.1778808 * |
MORDATCH ET AL.: "Robust physics-based locomotion using low-dimensionnal planning", ACM TRANSACTIONS ON GRAPHICS, vol. 29, no. 3, 2010 |
THIJS MANDERSLOOT ET AL: "Controlling Velocity In Bipedal Walking: A Dynamic Programming Approach", HUMANOID ROBOTS, 2006 6TH IEEE-RAS INTERNATIONAL CONFERENCE ON, IEEE, PI, 1 December 2006 (2006-12-01), pages 124 - 130, XP031053008, ISBN: 978-1-4244-0199-4 * |
VAN OORT, STRAMIGIOLI IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMOTION, 2007, pages 4653 - 4660 |
Also Published As
Publication number | Publication date |
---|---|
FR2981173A1 (fr) | 2013-04-12 |
EP2766251A1 (fr) | 2014-08-20 |
FR2981173B1 (fr) | 2013-11-29 |
US9317027B2 (en) | 2016-04-19 |
JP2014528368A (ja) | 2014-10-27 |
US20140257559A1 (en) | 2014-09-11 |
EP2766251B1 (fr) | 2016-12-28 |
JP6054404B2 (ja) | 2016-12-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Jenelten et al. | Perceptive locomotion in rough terrain–online foothold optimization | |
Siekmann et al. | Blind bipedal stair traversal via sim-to-real reinforcement learning | |
Feng et al. | Optimization‐based full body control for the darpa robotics challenge | |
Mordatch et al. | Ensemble-cio: Full-body dynamic motion planning that transfers to physical humanoids | |
EP2740012B1 (fr) | Robot a articulations de rigidite variable et methode de calcul de ladite rigidite optimisee | |
EP3568266B1 (fr) | Procédé de mise en mouvement d'un exosquelette | |
EP2448810B1 (fr) | Procede pour contrôler la marche d'un robot mobile et robot mettant en oeuvre le procede | |
Winkler et al. | Path planning with force-based foothold adaptation and virtual model control for torque controlled quadruped robots | |
Yi et al. | Online learning of a full body push recovery controller for omnidirectional walking | |
EP2766251B1 (fr) | Procédé de contrôle d'un robot possédant au moins deux jambes | |
EP4034344B1 (fr) | Procedes d'apprentissage de parametres d'un reseau de neurones, de generation d'une trajectoire d'un exosquelette et de mise en mouvement de l'exosquelette | |
Winkler | Optimization-based motion planning for legged robots | |
WO2022161297A1 (fr) | Procédé et appareil de commande de mouvement de robot, robot et support de stockage | |
WO2018042131A1 (fr) | Véhicule autonome compliant et procédé de commande associé | |
FR3117393A1 (fr) | Procédé de mise en mouvement d’UN exosquelette | |
WO2005062143A1 (fr) | Procede et dispositif de commande des deplacements d'une partie mobile d'un robot multi-axes | |
Chen et al. | Step ascent modelling of a two-wheeled robot | |
EP4103131B1 (fr) | Procedes de generation d'une trajectoire d'un exosquelette et de mise en mouvement de l'exosquelette | |
Xie et al. | Gait optimization and energy-based stability for biped locomotion using large-scale programming | |
Miller et al. | Improved stability of running over unknown rough terrain via prescribed energy removal | |
Lu et al. | Whole‐body motion planning and control of a quadruped robot for challenging terrain | |
Matsumoto et al. | The direction control of a biped robot using gyro sensor feedback | |
FR3139277A1 (fr) | Procédé de mise en mouvement stabilisée d’un exosquelette bipède | |
WO2024067440A1 (fr) | Procédé et appareil de commande de robot à deux roues, support, programme et robot | |
Saraf et al. | Convex Optimization in Legged Robots |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 12775799 Country of ref document: EP Kind code of ref document: A1 |
|
REEP | Request for entry into the european phase |
Ref document number: 2012775799 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2012775799 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 14350671 Country of ref document: US |
|
ENP | Entry into the national phase |
Ref document number: 2014535136 Country of ref document: JP Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |